108 ALASKAN GLACIER STUDIES a^ half mile, and at its head the undulating, moraine-covered ice appears, -with a depression between the two glaciers for several hundred yards further. On the south side of the valley rises the outer portion of the Variegated ice bulb, here low, undulating, and deeply covered with moraine. On the northern side (in 1905 and 1906) the-Hubbard Glacier rose more steeply, but still with low enough slope so that one could easily ascend it. Moraine completely covered this wall of the glacier and no ice was visible in it; but just north of it, on the Hubbard Glacier surface, the moraine became thinner and more crevassed, and in a short distance one came to the normal broken surface of the Hubbard. This moraine-covered margin extended to the sea and there rose as a debris-stained cliff, becoming rapidly higher and clearer of debris toward the west until the normal white ice cliff of the Hubbard front was reached. The amount of crevassing also increased away from the margin, and with it the amount of ice discharged. For a quarter of a mile from the glacier margin the debris-stained ice cliff was so inactive in 1905 and 1906 that iceberg discharge was practically absent. Only one or two small fragments were seen to fall, and the area of debris-stained cliff face was-far greater than the area whitened by recent falls. Moreover, there was pronounced-undercutting of the ice cliff by the sea water which proved clearly that iceberg discharge was not rapid. These facts are of importance by contrast with the condition in 1909 when an advance had begun along this part of the southeastern margin of the Hubbard Glacier. Hubbard Glacier Before Russell's Studies. In 1792 and 1794 Malaspina and Vancouver visited Yakutat Bay but tell us practically nothing specific concerning Hubbard Glacier. Malaspina's map, however, and the description by Malaspina and Vancouver were interpreted by Russell,1 Gilbert,2 and Davidson3 as proving that Hubbard Glacier, joined by Turner, then extended southward to a point south of Haenke Island and five or six miles south of where it ends now. The authors of this book have opposed this interpretation,4 because of evidence from existing vegetation and shorelines and absence of lacustrine deposits and of abandoned shorelines in a large part of Russell Fiord, and we place a different interpretation upon the conditions encountered by Malaspina in the light of existing floating ice phenomena, although we recognize the presence, rather recently, of a smaller lake in lower Russell Fiord. In Tebenkof's Atlas of 1848 the chart made, as Davidson has shown, by the Russian Booligin in 1807 and Lieut. Khromtchenko in 1823 has a crescentic ice front a short distance north of Haenke Island and a lake is shown in Russell Fiord. It is five or six miles long and there is no independent lake now. It is not dissimilar in width to upper Russell Fiord. It drains southward to the Pacific by the Situk river which is a little too long on the map and which now rises near the present head of Russell Fiord. The former-glacial lake in the head of Russell Fiord, whose shorelines we have studied, was of almost exactly this length and its outlet was probably the Situk. The front of Hubbard Glacier may have been north of Haenke Island in 1823 and the expanded Nunatak-Hidden Glacier reached the south part of Russell Fiord in that year i Russell, I. C., Nat. Geog. Mag., Vol. HE, 1891, pp. 67, 97-98; ISth Ann. Rept., U. S. Geol. Survey, 1894* p. 84. i Gilbert, G. K, Hkniman Alaska Expedition, Vol. m, 1904, p. 70. * Davidson, George, Trans, and Proc. Geog. Soc., Pacific, 1904, p. 51. < Tarr, R. S. and Martin, Lawrence, The Position of Hubbard Glacier Front in 1792 and 1794, Bull. Ameiv Geog. Soc., Vol. XXXIX, 1907, pp. 129-136.